Distributed Fusion Architectures and Algorithms for Target Tracking

Abstract

Modern surveillance systems often utilize multiple physically distributed sensors of different types to provide complementary and overlapping coverage on targets. In order to generate target tracks and estimates, the sensor data need to be fused. While a centralized processing approach is theoretically optimal, there are significant advantages in distributing the fusion operations over multiple processing nodes. This paper discusses architectures for distributed fusion, whereby each node processes the data from its own set of sensors and communicates with other nodes to improve on the estimates. The information graph is introduced as a way of modeling information flow in distributed fusion systems and for developing algorithms. Fusion for target tracking involves two main operations: estimation and association. Distributed estimation algorithms based on the information graph are presented for arbitrary fusion architectures and related to linear and nonlinear distributed estimation results. The distributed data association problem is discussed in terms of track-to-track association likelihoods. Distributed versions of two popular tracking approaches (joint probabilistic data association and multiple hypothesis tracking) are then presented, and examples of applications are given.

Uncertainty Analysis of Multiple Target Tracking in Distributed Sensor Networks

“…In addition, a distributed network can be also classified as the hierarchical and full distributed sensor network (DSW). The former is usually applied in the military domain, and it generally consists of three layers according to the data stream [16,17]: the data capture layer corresponding to sensor nodes, the data processing layer corresponding to local nodes, and the data analysis layer corresponding to the global node.…”

Decentralized State Initialization with Delay Compensation for Multi-modal Sensor Networks

“…However, this may not be possible in most cases even though the level sets do exist. Let α be any solution of (4). By the implicit function theorem [19], in the neighborhood of α, the level set L f (z m,t ) is an n − s dimensional manifold.…”

“…Particles can be generated by sampling uniformly from points in Λ and adding appropriate noise determined by the measurement model. Now consider the case, when s = n. By the inverse function theorem [19], given featuresẑ m,t , a unique inverse function f −1 m (·) exists in the neighborhood ofẑ m,t and therefore there exists a unique solution to (4) given by…”

Simulation Analysis on Optimal Guidance Law of Maneuvering Target Based on Multiple Model Adaptive Estimator

“…The weave guidance law is optimal in the sense that it requires the least acceleratio n against weave maneuvers [5]. The weave guidance law is shown to be…”